Disaggregated Quantization: Specializing LLM Prefill and Decode
Paper • 2609.26333 • Published • 47
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Disaggregated Quantization: Specializing LLM Prefill and Decode
GPTQ-2D: Cubic-Time Two-Sided Adaptive Rounding
BlackSamorez
authored 5
papers about 4 hours ago
Apertus LLM Family Expansion via Distillation and Quantization
Paper • 2605.29128 • Published
Bridging the Gap Between Promise and Performance for Microscaling FP4 Quantization
Paper • 2509.23202 • Published • 30
Apertus: Democratizing Open and Compliant LLMs for Global Language Environments
Paper • 2509.14233 • Published • 25
Unified Scaling Laws for Compressed Representations
Paper • 2506.01863 • Published • 19
I think might have had one too many experts stripped out
2
#11 opened about 8 hours ago
by 
Boffy
BlackSamorez
submitted a
paper to Daily Papers about 10 hours ago
A possible middleway
1
#8 opened about 13 hours ago
by 
mkristian
Is a safetensors release of GSQ-RCO planned? (needed for vLLM/ROCm)
1
#10 opened about 12 hours ago
by 
wolftrap101
IQ3_S
🔥 5
15
#26 opened 2 days ago
by 
DzmitryTheOtherOne
BlackSamorez
updated a
model about 14 hours ago
Will bf16 weights be available?
2
#5 opened 1 day ago
by 
tkyin
What is the SWE Verfied of the the pure GSQ RCO non coder Q3_S
1
#2 opened 2 days ago
by 
mayankiit04
Could we have a IQ3_S for coder?
🚀🔥 3
5
#3 opened 2 days ago
by 
alexaione
helcig
updated a
model 2 days ago
helcig
updated a
collection 2 days ago
helcig
published a
model 2 days ago
helcig
updated a
model 3 days ago
mkleinegger
updated a
model 8 days ago